1. Field of Invention
The present invention relates to means and methods for isolating a representative sample of a large quantity of particulate material such as coal.
Particulate material of relatively high value such as coal is frequently tested for properties or characteristics such as energy value (BTU per pound), metal content and/or moisture. Such tests are carried out on small samples of the subject material extracted from the primary shipment, source or supply. In the interest of accuracy and consistency, test samples are extracted and prepared with great care and methodical discipline to eliminate as many sources of weight, volume or moisture bias as economically feasible. The objective is a statistically "representative" sample wherein every particle in a sample source has equal opportunity to be in the selected volume.
For example, "representative" may be translated as a homogenous, 1000 g quantity of a 1000 ton shipment or flow increment. As produced in nature, the energy value of a coal seam is stratiated with some segments of greater heating value than others. However, with all the strata produced and combined, the seam product heating value and other properties may be relatively constant. Nevertheless it is necessary to test the product at shipment to determine its properties in relation to predetermined criteria, such as purchase contract specifications. The testing procedure must, of necessity, involve only small increments of the total quality of material. A proper test procedure must produce results which are confidently representative of the entire mass.
Pursuant to a commonly practiced methodology, a randomly selected cross sectional increment of coal or ore production as deposited on a conveyor is extracted completely from the conveyor. Depending on previous processing, the sample increment may be further pulverized and size screened. Commonly, this crushed and screened sample is divided into four identical test batches, one for the producer, one for the user and two held in abeyance for re-testing if necessary.
Many variables affect the momentary production rate that coal is deposited on a production conveyor, but the variance is manifest from the weight of coal on the conveyor per unit length of conveyor. Consequently, the batch magnitude of sample source extracted from the conveyor may fluctuate 200% or more. For example, the total sample source weight may vary from 25 pounds to 75 pounds of material.
Against this extracted sample source batch variation, the four, for example, formal samples to be isolated for actual testing are usually specified to be a small, uniform quantity such as 1 Kg., for example. Moreover, each formal sample may be specified as being composed of a designated number, 60, for example, of uniformly spaced, cross-sectional slices of the sample source batch. Meeting this specification of 60 increments for each of the four, 1000g formal samples, requires the sample source batch to be divided into 240 slices but only 16.6 g of each slice to be actually used for formal sample composition.
The prior art has developed both rotary and linear sample dividers by which a given sample source batch is distributed into a predetermined number of increments. None of these prior instruments, however, are capable of a randomly selected proportionalization of the sample source batch to achieve the necessary percentage of a specified number of increments in the specified number of samples to correctly allocate a variable weight source batch to a predetermined number of fixed weight formal samples.